Enhancing I0/I− Conversion Efficiency by Starch Confinement in Zinc–Iodine Battery

The redox couple of I0/I− in aqueous rechargeable iodine–zinc (I2‐Zn) batteries is a promising energy storage resource since it is safe and cost‐effective, and provides steady output voltage. However, the cycle life and efficiency of these batteries remain unsatisfactory due to the uncontrolled shuttling of polyiodide (I3− and I5−) and side reactions on the Zn anode. Starch is a very low‐cost and widely sourced food used daily around the world. “Starch turns blue when it encounters iodine” is a classic chemical reaction, which results from the unique structure of the helix starch molecule–iodine complex. Inspired by this, we employ starch to confine the shuttling of polyiodide, and thus, the I0/I− conversion efficiency of an I2‐Zn battery is clearly enhanced. According to the detailed characterizations and theoretical DFT calculation results, the enhancement of I0/I− conversion efficiency is mainly originated from the strong bonding between the charged products of I3− and I5− and the rich hydroxyl groups in starch. This work provides inspiration for the rational design of high‐performance and low‐cost I2‐Zn in AZIBs.


Introduction
Lithium-ion batteries have realized great commercial success in today's society and currently dominate the market for rechargeable batteries. [1,2]Nevertheless, cost and safety shortcomings hinder their extensive application in future, especially in large-scale grid-storage systems.[21][22] However, the key issues of I 2 fixation and interfacial electron conduction make I 2 -Zn battery performance unsatisfactory. [16,22,23]Most importantly, in aqueous electrolyte, I 2 molecules react very easily with I À to generate polyiodide (I 3 À and I 5 À ).The high solubility of polyiodide causes a shuttle effect, resulting in active material loss, capacity fading, and low coulombic efficiency. [3,24,25]he I 2 cathode usually relies on a conductive supporter (such as porous carbon) to transport electrons due to its insulating nature.[28][29] Besides carbon materials, other effective strategies have also been explored recently to support the I 2 cathode.For example, N-doped porous carbon, Prussian blue analog, and MXene were adopted as efficient hosts to suppress the loss of iodine and polyiodide. [16,22,25]Additionally, metal-organic frameworks were exploited as an ionic sieve membrane to restrict uncontrolled shuttling of triiodide and side reactions on the Zn anode. [30]However, these strategies using high-cost materials with complex fabrication procedures are not applicable for large-scale production of cost-efficient I 2 -Zn batteries.
Starch, on the contrary, is a very low-cost and widely sourced food used daily around the world.Inspired by the classic chemical phenomenon in which starch turns blue when it encounters iodine, here we found that the I 0 /I À conversion efficiency of an I 2 -Zn battery is drastically enhanced by starch confinement.The mechanism is mainly based on the strong bonding between the charged products (I 3 À and I 5

À
) and the rich hydroxyl groups in starch.The I 2 -Zn battery using carbon cloth as cathodic substrate and ZnI 2 -starch (ZIS) as electrolyte achieved much higher charge/discharge efficiency and capacity than that in ZnI 2 (ZI) solution.Benefiting from the ZIS gel electrolyte in I 2 -Zn battery, the capacity still retained nearly 92% after 2000 cycles.Additionally, by using a constant voltage charge method, fast battery charge and The redox couple of I 0 /I À in aqueous rechargeable iodine-zinc (I 2 -Zn) batteries is a promising energy storage resource since it is safe and costeffective, and provides steady output voltage.However, the cycle life and efficiency of these batteries remain unsatisfactory due to the uncontrolled shuttling of polyiodide (I 3 À and I 5 À ) and side reactions on the Zn anode.
Starch is a very low-cost and widely sourced food used daily around the world."Starch turns blue when it encounters iodine" is a classic chemical reaction, which results from the unique structure of the helix starch molecule-iodine complex.Inspired by this, we employ starch to confine the shuttling of polyiodide, and thus, the I 0 /I À conversion efficiency of an I 2 -Zn battery is clearly enhanced.According to the detailed characterizations and theoretical DFT calculation results, the enhancement of I 0 /I À conversion efficiency is mainly originated from the strong bonding between the charged products of I 3 À and I 5 À and the rich hydroxyl groups in starch.This work provides inspiration for the rational design of high-performance and low-cost I 2 -Zn in AZIBs.
superior rate performance were achieved, and there was no obvious decay for the capacity unless the current density reached 10 mA cm À2 .Moreover, the entire battery is constructed only from an ordinary carbon substrate, ZIS gel electrolyte, and a Zn anode, with no additional cathode materials or complex procedures needed.Thus, the starch confinement strategy will inspire the development of simply constructed and low-cost I 2 -Zn aqueous batteries.Starch, which is very low cost and is widely sourced from wheat, corn, potatoes, etc., has a typical helix structure as shown in Figure 1a.As previously reported, the rich hydroxyl groups in the helix structure easily bond with I 2 molecules to generate a starch-iodine compound in blue color. [31,32]The I 2 -Zn battery was assembled as shown in Figure 1b, with pure carbon cloth serving as the cathodic substrate, ZIS gel serving as the electrolyte, and Zn foil serving as the anode electrode.A photograph of ZIS gel is shown in Figure S1, Supporting Information.Here, the ZIS gel is also the source of iodine, which serves as raw cathode material, so no further complex procedures are required to assemble the I 2 -Zn battery.Scanning electron microscopy (SEM) images of the pure carbon cloth electrode and carbon cloth@ZIS electrolyte are displayed in Figure S2a The percentage of iodine on the carbon cloth electrode is 30.6 wt.% as detected by the EDS spectrum shown in Figure S3, Supporting Information.Additionally, after drying, pure ZIS electrolyte exhibits a smooth surface as shown in the SEM image displayed in Figure S4a, Supporting Information.The corresponding EDS mapping of elemental C, O, Zn, and I in Figure S4b,e, Supporting Information, again reflects that ZnI 2 is uniformly distributed throughout the starch in ZIS.As shown in Figure S4f, Supporting Information, the percentage of iodine in pure ZIS electrolyte is around 38 wt.%.Ordinary particles were observed in the ZIS gel after drying as shown by the transmission electron microscopy (TEM) images in Figure S5a,b, Supporting Information, while the elemental I distribution in such particles is shown by the corresponding EDS mapping in Figure S5c, Supporting Information.

Electrochemical Assessment
By introducing starch in the electrolyte, high capacity and ultrafast charge were realized by using a constant voltage charge method with a potential of 1.4 V (Figure 2a) in ZIS electrolyte.The discharge time at the current density of 1 mA cm À2 reached 1500 s, much longer than the charge time of 300 s, illustrating that the constant voltage charge mode can greatly shorten the charge time.The theoretical voltage of an I 2 -Zn battery is 1.38 V versus Zn/Zn 2+ .Here, the discharge potential of the cell reached 1.28 and 1.32 V at the charge potentials of 1.4 and 1.6 V, respectively, which are very close to its theoretical voltage.The insignificant overpotential reveals the low energy barrier and fast conversion of I 0 /I À redox couples.However, the discharge capacity and output voltage are drastically decreased in pure ZnI 2 (ZI) solution.The spike observed in the corresponding discharge curves (marked by dashed red circles in Figure 2a) illustrates the poor reversibility and unstable electrochemical reactions of I 0 /I À conversion.To visually understand the I 0 /I À conversion processes, photographs of an I-Zn battery in ZI solution in a beaker were taken after various charge time and are shown in Figure S6a-c, Supporting Information.
Clearly, the generated iodine dissolved and rapidly diffused throughout the aqueous ZI electrolyte, which resulted from the intrinsic solubility of polyiodide.After 300 s charge time, the solution was dripped into starch, starch immediately turned blue as shown in Figure S6d, demonstrating that zero-valent iodine was distributed throughout the ZI electrolyte.The poor performance and the in situ observation of the charge process of the I-Zn battery in ZI solution confirm that its low I 0 /I À conversion efficiency directly results from the loss of iodine on the carbon electrode.At a charge potential of 1.4 V for the first cycle, the capacity and coulombic efficiency of the I-Zn battery in ZIS and ZI electrolyte are shown in Figure 2b.Notably, the I 2 -Zn Energy Environ.Mater.2024, 7, e12522 battery in ZIS electrolyte achieved superior capacity and coulombic efficiency compared to that in pure ZI, illustrating that the I 0 /I À conversion efficiency is drastically enhanced by starch confinement.The cyclic voltammetry (CV) curves of ZIS were further investigated as shown in Figure S7, Supporting Information.The increasing current above 1.25 V is exactly attributed to the oxidation of I À , while the wide reduction peaks ranged from 1 to 1.25 V correspond to the conversion of I 0 to I À .The conductivity of the ZIS gel is also tested as shown in Figure S8, Supporting Information, and the ionic conductivity of the ZIS gel was calculated to be about 2.7 9 10 À5 S cm À1 .Figure 2c summarizes the first discharge capacity of the I 2 -Zn battery in ZIS electrolyte at various charge potentials (1.4,1.6, and 1.8 V) and time (2, 5, 10, and 30 min).The corresponding first charge/discharge curves and columbic efficiencies are shown in Figures S9-S11, Supporting Information.Capacity was found to gradually increase with increasing charge time for all potential conditions.The maximum capacity reached 4.1 mAh cm À2 at 1.6 V after 30 min charge time.However, the I 2 -Zn battery in ZI electrolyte had much worse capacity and coulombic efficiency as shown in Figure S12, Supporting Information, again demonstrating that I 0 /I À conversion efficiency and discharge capacity are clearly optimized by starch confinement.It is worth noting that, in ZIS electrolyte, neither the capacity nor the coulombic efficiency benefit from the highest charge potential of 1.8 V. Presumably, the iodine loss resulted from diffusion is serious at too high potential.Since realistic battery applications should have both high capacity and high coulombic efficiency, we think the most appropriate conditions are potential of 1.4 or 1.6 V and 2 or 5 min charge time.Rate performance of the I-Zn battery in ZIS electrolyte at the potential of 1.6 V for 5 min charge time is shown in Figure 2d.Impressively, the capacity is around 1.4 mAh cm À2 and has no decrease even at the high current density of 10 mAh cm À2 , proving the battery's fast I 0 /I À conversion ability and superior electrochemical activity.It was found that the coulombic efficiency rises sharply after the first charge/discharge cycle, indicating that I 0 /I À conversion is quickly activated after the first charge/discharge stage.The coulombic efficiency remained at around 80%-90% throughout subsequent cycles.Unusually, it was found to be higher than 100% when the current density returned to 1 mA cm À2 , which resulted from the accumulated unreacted iodine from the previous cycles.The potential of 1.4 V and charge time of 5 min are selected as the representative condition to perform the cycle stability for its relatively high capacity and superior columbic efficiency, and the corresponding discharge current density is 10 mA cm À2 .As shown in Figure 2e, the capacity was 0.78 mAh cm À2 after several cycles of initial activation, and the battery retained nearly 92% capacity over 2000 cycles.The corresponding discharge curves at the 1st, 200th, 500th, 1000th, and 2000th cycles are shown in Figure S13, Supporting Information.It is also worth noting that the coulombic efficiency increases from 70% to 95% after 200 cycles of initial activation and is well maintained over 2000 cycles.The SEM image of the cathode after cycling shown in Figure S14a, Supporting Information, and the corresponding EDS spectrum shown in Figure S14b, Supporting Information, indicate that there is still a high iodine content of 35.5 wt.% compared with that in the initial pure ZIS electrolyte (38 wt.%), demonstrating that there is nearly no iodine loss or diffusion even after 2000 cycles.The highest capacity of 4.1 mA cm À2 and the average capacity of around 0.8 mAh cm À2 with high columbic efficiency of 95% surpass most of the recent reports in I-Zn batteries (Figure S15, Supporting Information). [16,23,27,28,30,33,34]dditionally, galvanostatic charge/discharge and the rate performance at various current densities were performed as shown in Figure S16a,b, Supporting Information.The results reflect the superior stability of ZISbased I-Zn battery, especially at high current densities.Then, the galvanostatic charge/discharge cyclic test was also conducted as shown in Figure S17, Supporting Information.The capacity reached 0.83 mAh cm À2 at high current of 10 mA cm À2 , and both the capacity and columbic efficiency kept well within 1200 cycles.Energy Environ.Mater.2024, 7, e12522

Mechanism Exploration of Starch Confinement
The function of the starch confinement in improving the I 2 -Zn battery performance was explored through some characterizations.First, Raman spectra of the cathodes in the ZI and ZIS electrolytes were collected after charging (Figure 3a).The I 3 À and I 5 À peaks located at 98, 210, and 151 cm À1 for the ZIS electrolyte show clearly lower wavenumbers than those for the starch-free ZI electrolyte (138, 236, and 164 cm À1 ), [16,24,25,[35][36][37] reflecting the lower Raman scattering energy of the I 3 À and I 5 À species in ZIS electrolyte and further proving that the I 3 À and I 5 À species are bonded with the starch.[40] X-ray photoelectron spectrometry (XPS) was employed to analyze the conversion Energy Environ.Mater.2024, 7, e12522 mechanism of I À as shown in Figure S18, Supporting Information, the peaks of I À shift to higher binding energy due to its oxidation to I 0 . [16]hen, the XPS spectra of I with and without starch in the electrode after charging were further examined (Figure 3b).The binding energy of the peaks located at 618.8 and 630.4 eV for the ZIS electrolyte is slightly lower than that for the ZI electrolyte (619.8 and 631.2 eV), [41,42] which proves that the electrons in the ZIS electrolyte transfer from the starch to the iodine.To further analyze the bonding between I and starch, a comparison between high-resolution XPS spectra of carbon for the ZIS electrolyte after charging and pure starch is provided in Figure 3c.Interestingly, both the intensity and ratio of the C-OH bond peak (286.4 eV) in the spectrum for the ZIS electrolyte after charging are much lower than those of the corresponding peak for pure starch, [43,44] indicating that the C-OH bond is weakened by the introduction of polyiodide (I 3 À and I 5 À ) and thus indirectly confirming that the polyiodide is bonded with hydroxy (OH) groups in the ZIS electrolyte after charging.To further verify this, Fourier transform infrared (FTIR) spectroscopy was performed for pure starch and the ZIS electrolyte after charging, and the results are displayed in Figure 3d. [45,46]With reference to the O-H vibration for pure starch (1650 cm À1 ), that for discharged ZIS visibly shifts to a lower wavenumber due to the bonding between polyiodide and the abundant OH in starch.Influenced by this, both the C-O (1154 cm À1 ) and C-O-H (1075 cm À1 ) bonds also shift slightly to lower wavenumbers.Thus, the generated polyiodide species (I 3 À and I 5

À
) are believed to be bonded with the OH group in starch and to weaken the related C-O and C-O-H bonds, corresponding to the interaction between hydroxyl group and carbon in hexatomic carbon ring.On the contrary, there is nearly no change for the other bonds, such as C-C at 1012 cm À1 , C-H at 2914 cm À1 , and skeleton vibration below 1000 cm À1 , since they are far away from the I-OH bond.Analyses of the Raman, XPS, and FTIR spectra all indicate that the function of starch confinement results from the bonds formed between polyiodide species and OH in starch.As schematically illustrated in Figure 3e, after cycling, the ZIS electrolyte near the Zn anode was stripped to perform FTIR spectroscopy.Notably, as shown in Figure 3f, it was found that there is nearly no shift for any of the peaks in the spectrum for the ZIS electrolyte after cycling compared with that for pure starch, which demonstrates that starch can limit the shuttling and diffusion of polyiodide from the carbon substrate toward the Zn anode.XRD patterns, XPS spectra, and SEM images of Zn anode after cycles are also shown in Figures S19 and S20, Supporting Information.XRD results (Figure S19a, Supporting Information) show that ZIS can protect the Zn anode and avoid the formation of by-products of Zn (IO 3 ) 2 .Then, XPS spectra verify that the I intensity on Zn surface with ZIS gel is much lower than that for ZI electrolyte.Finally, SEM images directly prove the dendrite growth of Zn anode in pure ZI electrolyte.In contrast, Zn surface only shows smooth morphology in ZIS gel electrolyte.Thus, FTIR, XRD, XPS, and SEM characterizations all prove that starch can hinder the shuttling and diffusion of polyiodide from the carbon substrate toward the Zn anode.Additionally, the Zn/Zn symmetric cells were also assembled in ZI and ZIS electrolyte, respectively, to further verify the protection ability of ZIS on Zn anode (Figure S21, Supporting Information).The Zn/Zn symmetric cell worked stably in ZIS gel electrolyte for 100 h at the current density of 2 mA cm À2 , which significantly surpasses that in ZI electrolyte.Even at the high-rate charge/discharge current (5 mA cm À2 ), the cycle life of this cell in ZIS electrolyte still surpasses 200 h.

Theoretical Research and the Summary of Starch Confinement Mechanism
To the best of our knowledge, no detailed studies of the bonds between triiodide/pentaiodide and starch have been reported thus far.[49] To simplify the calculation model, we used a ring molecule constructed from six glucose molecules to represent the helix structure of starch molecule.We first calculated the adsorption energy of traditional I 2 molecule-starch compounds.The adsorption energy of À1.011 eV using six glucose molecules is much lower than that for one glucose molecule (À0.182 eV), indicating that starch, with a ring of polyglucose molecules, exhibits a much more powerful ability to bond with iodine than a single glucose molecule.Thus, the abundant OH groups in starch are key for confining the shuttling and diffusion of iodine.Subsequently, the models of starch and polyiodide (I 3 À and I 5

À
) were established to calculate their adsorption energy.Notably, the adsorption energy values of both I 3 À (À4.550 eV) and I 5 À (À4.751 eV) are much lower than that of the I 2 molecule, which means that the bonds between polyiodide and starch are stronger than traditional I 2 molecule-starch bonds.The extremely low adsorption energy below zero indicates that the formation of the binding is very feasible in thermodynamics.Thus, the DFT calculation results also imply that the high I 0 /I À conversion efficiency and high capacity originate from polyiodide-hydroxyl bonds formed by starch confinement.Based on the above experimental investigation and theoretical analysis, the starch confinement mechanism is clearly summarized through schematic illustrations shown in Figure 5.During the charge process (Figure 5a), I À ions around the carbon cloth substrate are first oxidized to become polyiodide (I 3 À and I 5 À ), and shuttling and diffusion of the polyiodide species are then restricted by the abundant OH groups in starch.Meanwhile, Zn 2+ ions are deposited on the Zn anode.During the subsequent discharge process (Figure 5b), the generated ) is again reduced on the carbon cloth substrate and the corresponding anode Zn metal is oxidized to become Zn 2+ ions.In this simply constructed I 2 -Zn battery, carbon cloth serves as a superior conductive substrate that promotes the ultrafast I 0 /I À conversion and then starch provides a powerful barrier to confine the shuttling and diffusion of the polyiodide.This synergistic effect between carbon and starch leads to high degrees of I 0 /I À conversion efficiency, rate performance, and capacity.

Conclusion
In summary, inspired by the classic chemical reaction in which starch turns blue when it encounters iodine, we proposed a starch confinement strategy to restrict the shuttling and diffusion of the polyiodide in an I 2 -Zn battery.In ZIS electrolyte, this battery was found to possess high degrees of area capacity, stability, and I 0 /I À conversion efficiency.Detailed experimental investigations, including Raman, XPS, and FTIR analyses, as well as DFT calculations, confirmed that the starch confinement originates from the strong bond between polyiodide (I 3 À and I 5

À
) and the abundant hydroxyl groups in starch.With the help of ZIS electrolyte, the I 2 -Zn battery achieves a high capacity of 4.1 mAh cm À2 and superior rate performance without decay even at the high current density of 10 mA cm À2 .It also exhibits a long cycle lifespan, and the capacity retains nearly 92% capacity over 2000 cycles.The simply constructed and inexpensive I 2 -Zn battery incorporating starch is promising for future practical I 2 -Zn batteries.

Experimental Section
Synthesis of ZIS electrolyte and battery assembly: ZnI 2 -starch (ZIS) gel electrolyte was fabricated by mixing starch, ZnI 2 , and deionized water.First, 3.19 g ZnI 2 (Macklin Biochemical Co.) was dissolved in 10 mL deionized water in a beaker to fabricate 1 M ZnI 2 solution as ZI electrolyte, and then 2 g commercial starch was dissolved in the ZI solution at a temperature of 80 °C by constant stirring for 30 min to prepare ZIS gel electrolyte.The battery was then assembled as a pouch cell with carbon cloth serving as the cathode substrate, a moderate amount of ZI solution or ZIS gel serving as the electrolyte and a piece of Zn foil serving as the anode.The size of the cathode and anode inside ranged from 0.8 to 1.2 cm À2 , and the gel volume was controlled to be 200 lL in the pouch cell with mass of 240 mg.Material characterization and electrochemical measurements: The morphology and microstructure of the prepared samples were characterized by field-emission scanning electron microscopy (FE-SEM, JSM-6700F) and transmission electron microscopy (TEM, JEM-2100F).The elemental composition was determined by SEM energy-dispersive X-ray spectroscopy (SEM-EDS, ZEISS Supra 55 and Oxford X-Max), for which the corresponding acceleration voltage was 20 kV.Raman spectra were collected using a Raman spectrometer (LabRAM HR Evolution, 532 nm).X-ray photoelectron spectroscopy (XPS) was performed using a PHI Quantera SXM scanning X-ray microprobe to analyze the binding energy of the elemental I and C. Fourier transform infrared (FTIR) spectrometry (Nicolet iS50) was performed to determine the chemical group and absorption wavenumber of the starch and ZIS.All electrochemical measurements of the I 2 -Zn battery, including charge/discharge testing, rate performance, and cyclic lifespan, were performed using a battery testing system (Landt CT2001A, China).Density functional theory (DFT) calculations: All calculations were performed using Vienna Ab initio Simulation Package (VASP) software.The projector augmented wave (PAW) method was used to describe the electron-ion interaction.The exchange and correlation effects of the electrons were described by a local-density approximation (LDA) method, and spin polarization was considered.The cutoff energy of the plane wave was set as 400 eV.The (2 9 1 9 2) k-point mesh was used for k-space integration in our structure relaxations.A conjugate-gradient algorithm was used to relax the ions into their instantaneous ground state, and partial occupancies were set for each orbital with Gaussian smearing.The structures involved were fully relaxed with energy and force convergences of less than 1 9 10 À6 eV and 0.03 eV A À1 , respectively.

2. 1 .
Proposal of the I 2 -Zn Battery with Starch Confinement ,b, Supporting Information and Figure 1c.It can be seen that the carbon cloth is fully covered by the ZIS electrolyte.The corresponding energydispersive X-ray spectroscopy (EDS) mapping images (Figure 1d-g) show the uniform distribution of elemental C, O, Zn, and I, revealing the coexistence of starch and ZnI 2 in the ZIS electrolyte.

Figure 1 .
Figure 1.a) Various sources of starch and top and side views of its helix structure.b) Assembled I 2 -Zn battery structure.c) SEM image of carbon cloth@ZIS electrolyte and corresponding EDS mapping of elemental d) C, e) O, f) Zn, and g) I.

Figure 2 .
Figure 2. a) Constant potential charge and galvanostatic discharge curves in ZI or ZIS electrolyte and at different potentials.b) Capacity and coulombic efficiency in ZI or ZIS electrolyte and at charge potential of 1.4 V as functions of charge time.c) Summary of discharged capacity for various potentials and charge time in ZIS electrolyte.d) Rate performance in ZIS electrolyte at potential of 1.6 V for 5 min charge time.e) Cyclic lifespan investigation of the I 2 -Zn battery in ZIS electrolyte at the potential of 1.4 V for 5 min charge time, where the discharge current density is 10 mA cm À2 .

Figure 3 .
Figure 3. a) Raman and b) high-resolution XPS spectra of I in ZI and ZIS electrolytes on carbon cloth substrate after charging.c) High-resolution XPS spectra of C and d) FTIR spectra for pure starch and ZIS electrolyte after charging.f) FTIR spectra for ZIS electrolyte near the Zn anode removed after cycling, as shown in the schematic illustration displayed in e), and for pure starch.